https://open-source-silicon.dev logo
Title
a

Art Scott

04/29/2022, 12:08 AM
Next Generation Arithmetic Conference on Next Generation Arithmetic (CoNGA) is the leading conference on emerging technologies for computer arithmetic. ‘The impact of the next-generation arithmetic is similar to that of a turn of Moore’s law – doubling computing capability but without having to shrink transistors. Current floating-point numbers (“floats”) were standardised in the 1980s when transistors were millions of times more expensive than they are now; floats make compromises that are overdue for update. The ability to train neural networks with fewer bits (≤16 bits) provides an enormous performance boost for artificial intelligence (AI), presently the most important computing focus and challenge. Novel arithmetic is a game changer for the computing industry. The game in High Performance Computing (HPC) presently is to achieve exascale computing, which most practitioners implicitly assume means a billion double-precision floating-point operations per second while meeting the critical power envelope challenge of staying under 20 MW. The ability of the new generation arithmetic to use fewer bits for the same/higher computation accuracy will facilitate meeting both the performance and power challenges of exascale computing. Unum arithmetic, in particular the newest type of unums (posits), is an example of a disruptive new format. Posits were designed at the request of the RISC-V committee. They have higher accuracy and a larger dynamic range using the same number of bits, or what is perhaps more disruptive, sufficient accuracy using fewer bits, thereby saving storage, bandwidth, energy and power. Posits produce bitwise-identical results across diverse computer systems; current floats cannot even guarantee identical results on successive program runs. To a large extent, posits can serve as drop-in replacements for floats of the same precision; however, the advantages of increasing speed while lowering chip area, power/energy consumption, storage, and bandwidth demands come from using posits to reduce precision requirements without sacrificing accuracy. Changing arithmetic however affects the entire hardware-software stack from chip design to the application. The Conference on Next Generation Arithmetic provides an excellent opportunity for computer arithmetic enthusiasts to gather, interact and exchange ideas on not only what the next generation arithmetic should be, but also the news and updates on the latest developments of breakthroughs with next generation data formats and their corresponding hardware, tools, applications, services, etc..’ Stanford Seminar, Beyond Floating Point: Next Generation Computer Arithmetic https://www.linkedin.com/feed/update/urn:li:activity:6823778113026174976/ “The posit number format is proving ideal for Machine Learning and is now in use by dozens of institutions, both commercial and academic. … it is not beholden to any commercial vendor and utterly non-proprietary…” Type III Unum (Posit & Valid) and Implementations https://en.wikipedia.org/wiki/Unum_(number_format)#Type_III_Unum_(Posit_&_Valid)